[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: L4-hurd Digest, Vol 6, Issue 12
From: |
Niels Möller |
Subject: |
Re: L4-hurd Digest, Vol 6, Issue 12 |
Date: |
22 May 2003 10:51:31 +0200 |
User-agent: |
Gnus/5.09 (Gnus v5.9.0) Emacs/21.2 |
Niklas Soderlund <address@hidden> writes:
> > To get a multiserver system as fast as a well-crafted monolithic
> > kernel (like bsd or linux) seems quite hard.
> But possible?
Don't know. I guess the L4 folks believe so, at least. For l4-hurd, a
reasonable performance goal is probably to be not much slower than
Linux or BSD.
> Ok, I have always thought that microkernel-based systems were faster than
> monolithickernel systems.
> I am new to this, could you please explain some benefits with microkernel
> against monolithic?
For the Hurd, the primary reason for using a micro-kernel is that it
makes it possible to move most of the features traditionally
implemented in the kernel to user space servers, and to allow ordinary
users to replace some or (almost) all of these servers. See
http://www.gnu.org/software/hurd/hurd-paper.html.
/Niels
- Re: L4-hurd Digest, Vol 6, Issue 12, Fortes Marcelo, 2003/05/16
- Re: L4-hurd Digest, Vol 6, Issue 12, Marcus Brinkmann, 2003/05/17
- Re: L4-hurd Digest, Vol 6, Issue 12, Niklas S!P0, 2003/05/19
- Re: L4-hurd Digest, Vol 6, Issue 12, Niels Möller, 2003/05/20
- Re: L4-hurd Digest, Vol 6, Issue 12, Niklas Soderlund, 2003/05/21
- Re: L4-hurd Digest, Vol 6, Issue 12,
Niels Möller <=
- Re: L4-hurd Digest, Vol 6, Issue 12, Ludovic Courtès, 2003/05/22